Disentangling the effects of geographic distance, environment and history on beta diversity of freshwater fish at a biogeographical crossroads.

ABSTRACT

Examining assemblage turnover and variation along geographic and environmental distances is a useful approach to evaluate beta diversity patterns and associated driving mechanisms. However, such studies are relatively limited in freshwater systems. Here, we compared the relationships between freshwater fish beta diversity and geographic distances among 165 hydrological units (HUs) in four zoogeographical regions (PA, Palearctic Region; CA, High Central Asia; EA, East Asia, SA, South Asia) across China and adjacent areas. This area can be considered a biogeographical crossroads, where faunal composition shares elements with different biogeographic and evolutionary origins. We found a considerably high level of between-HU overall dissimilarity (ßsor, range from ca. 0.60 to 0.85) in all four regions, mainly due to the turnover component (the relative contribution of ßsim to ßsor ranged from 60% to 90%). In general, ßsor and ßsim both significantly increased with geographic distance (except in PA), whereas the nestedness-resultant component (ßsne) decreased with geographic distance. The intercepts and slopes of the relationships between dissimilarities and distance (RDDs) both varied significantly among the four regions. The intercepts of ßsor and ßsim were both highest in SA, followed by CA, PA and EA, implying different levels of fish faunal heterogeneity at short distances. In contrast, the slopes of these two dissimilarities followed the decreasing trend from EA > CA > SA > PA, suggesting different environmental suitability and dispersal ability of fish species among regions. Variation partitioning in distance-based redundancy analysis showed that the spatial and historical factors were more important than area-heterogeneity and energy factors across all HUs and within three individual ecoregions (EA, SA and CA), but spatial factors were non-significant in PA. Our study highlighted the usefulness of RDDs in understanding biogeographical patterns and enhancing the biodiversity conservation of freshwater fishes.